Summary The Canadian Coast Guard Messerschmitt-Bolkow-Blohm (MBB) BO105 helicopter (registration C-GCHX, serial number S695), operated by Transport Canada as CTG357, was carrying out external load operations near Bella Bella. It had completed 27external loads and was returning to the Canadian Coast Guard ship (CCGS) Bartlett from Dryad Point Lighthouse Station with an empty cargo bonnet. En route over the water at an altitude of about 200feet, the bonnet went above and behind the tail rotor, and the longline hung up on the back of the helicopter. The helicopter slowed, began to descend, turned right and then crashed into the water. It sank immediately. The pilot was able to exit the sunken helicopter but remained face down in the water. He was wearing an uninflated lifejacket. The pilot was rescued within three minutes and revived, but remained in critical condition for several days. The helicopter was found at a depth of 26metres on an ocean floor slope. Ce rapport est galement disponible en franais. Other Factual Information The weather was as follows: scattered cloud at 2000feet, visibility about 15miles, winds calm, and the seas rippled. The sling equipment being used was a 33-metre synthetic rope line (longline) attached to the helicopter's belly hook, with a swivel hook at its end. Attached to that extended hook was a bonnet (see Photo1) with some rope gear and slings inside. It was lashed closed with a polypropylene rope. The bonnet weighed less than 10kg, and the total external load weight was less than 40kg. The surface area of the external load was about 2.6square metres. Photo3 shows the actual gear and its configuration. The pilot was flying from the right seat with the right door removed. The pilot was wearing the lap belt; however, he was not using the available upper-body restraint system, as it restricts pilots from leaning out to view the external sling load. The helicopter was moving at about 60knots when the pilot felt and noticed the bonnet open. The sling gear was still in it, but the bonnet was folded in (see Photo1) between the one o'clock and the two o'clock position. About two seconds later, the bonnet and longline travelled to the port side of the helicopter, appearing at the pilot's ten o'clock position, slightly above eye level. It moved backwards and out of sight, then there was a bang followed by some thrashing noises, and the helicopter began to turn right. The pilot could see the line hanging from high on the tail, trailing down, back and out at about the four o'clock position. The bonnet was not evident on the line. As this was happening, the pilot slowed the helicopter and reduced power by lowering the collective and bringing the throttles to the engine idle position. Just before impact, the pilot increased the throttles to full and raised the collective a small amount. The helicopter hit the water sooner than expected, and the pilot had not activated the emergency floatation system. The helicopter sank immediately after impact with the water. The pilot was wearing a helmet, and even though it was structurally damaged from impact, he remained conscious as the helicopter sank. The pilot unlatched his lap belt and swam to the surface. He did not remember having a life vest on and did not inflate it. At the surface, the pilot tried to grab items to hold onto, but they sank and he could not keep his head above the water. A boater saw the accident and was picking up some of the debris when he noticed the top of the pilot's helmet. The helmet was painted grey and the pilot was wearing a navy blue flight suit, which was hard to see in the ocean. The uninflated yellow floatation device was protected by a navy blue life-vest cover and thus not readily visible. As the boater got closer to the helmet, he recognized it as a helmet with a body attached. The boater was able to get the pilot's head out of the water until Canadian Coast Guard rescue boats arrived and attendants revived the pilot. Photo2. Bonnet showing becket and separation The helicopter wreckage was photographed and the sling gear was recovered. The longline was wrapped around the tail-rotor driveshaft, between the tail rotor and the 90 gearbox. The line had released from the helicopter's belly hook and lay about 2metres away from the helicopter hook. The line was broken about 8metres from its lower hook. All the beckets of the bonnet (see Photo2) were secured in a short sling, which was secured in the lower hook. The polypropylene rope was still wrapped around the beckets, but it was loose (see Photo3). Two of the beckets were partly separated, one about 23cm from the lip of the bonnet (see Photo2). Further examination of the helicopter wreckage on land revealed that the longline was released before the helicopter struck the water. It was determined that persons seated in either of the front seats were able to hit their heads on a fixture between and behind the two front seats, installed to hold a liferaft. Only the liferaft support structure is near enough to be contacted by the front-seat occupants' heads, even when the occupants are secured in their seats with the available upper-body restraint devices. This modification was installed by the operator in accordance with an approved Limited Supplementary Type Certificate. The pilot was trained and certificated appropriately. He had extensive experience in external load operations and had completed underwater egress training. Records indicate that the helicopter was maintained to the required standards. In 1993, the TSB carried out an engineering analysis of helicopter sling aerodynamics and produced Engineering Report LP 51/93. At that time, in response to the engineering report and occurrence A92W0177, the Board issued recommendationA93-12, directed to Transport Canada. The recommendation highlighted the fact that, within the industry, there had been several helicopter accidents resulting from tail-rotor damage caused by contact with slings. Those accidents resulted in several fatalities, serious injuries and, in most cases, substantial aircraft damage. A large portion of the accidents involved flights with an intentionally empty, unweighted sling, even though the helicopter community recognized this to be a hazardous practice. It was recommended that Transport Canada coordinate the development and implementation of airworthiness standards and operational limitations for helicopter slinging equipment. The response from Transport Canada indicated that the responsibility to ensure safe slinging would remain with operators, with a requirement that company operations manuals would include instructions on the type of equipment to be used for specific operations, procedures, equipment inspection and maintenance. The use of slinging equipment in an unsafe manner has continued, as demonstrated by this occurrence. In 2001, the TSB Director of Investigations (Air) drew to Transport Canada's attention, in the form of Safety AdvisoryA010006, that although the seats in the cockpit of helicopters were equipped with four-point safety harnesses, pilots used only the lap-belt portion of the harness. This was only loosely fitted when carrying out vertical reference slinging operations, and the equipment provided is ill-suited to the task of vertical reference flying during external load operations. Moreover, an analysis of helicopter crash dynamics by Coltman1 showed that, of the personnel who experienced a helicopter crash, only 9percent of those who were wearing a shoulder harness had severe injuries, compared with 34.3percent of those who wore only a lap belt. Transport Canada's response to the safety advisory letter was that it was the industry's responsibility to comply with the regulations and, if warranted, apply for an approval of a configuration to meet its operational needs. The practice of operating without the use of upper-body restraint equipment continues, as demonstrated in this occurrence. Slinging external loads using vertical reference techniques is common in helicopter operations around the world, and very common in Canada. Most helicopters are not designed to accommodate this practice, and certification for external load operations does not take these techniques into account.